Fax confirming a San Antonio “school-bus ghost tracks” story
Host Art Bell recaps a story he says was told the previous hour by [quote withheld] It was about haunted railroad tracks where children in a school bus were killed. A stopped vehicle there is said to move by itself across the rails, leaving children’s handprints in baby powder put on cars. Bell then reads a fax from someone signing as [quote withheld] The writer says the story is true and was on their local news that day, with video of the car moving on its own and of the fingerprints. Bell adds no further comment in this part of the broadcast.
This clip comes from the source recording. AI chose where it starts and ends, so it may start or end a little early or late. The word-for-word transcript is not published. You can hear the source recording on the Internet Archive (file “Ghost to Ghost 1996.mp3”, starting at 1:30:58). Is this you, or someone you know? Ask for a correction or removal.
The story as told
Host Art Bell, introducing the item, says that in the previous hour [quote withheld] told [quote withheld] about haunted railroad tracks. In his recap, children in a school bus were killed by a train on the tracks. “The story goes” that since then, anyone who drives near the tracks and stops their vehicle finds it moving by itself and being pushed across. He adds that people have been putting baby powder on the backs of their cars and finding “children’s handprints” in it. Bell then says he has [quote withheld] and reads it. The writer asks whether Bell knows [quote withheld] and states “it’s true.” It was on one of their local news broadcasts that day, which [quote withheld] and [quote withheld] The fax is signed [quote withheld] who adds [quote withheld] Bell makes no further comment in this part of the broadcast, and the next caller follows. Nobody in this part of the broadcast claims to have seen the events. The story is Bell’s summary of another caller plus a fax that itself cites a television news report.
Reported events, in order
- 0:00
Art Bell recaps that [quote withheld] told a story the previous hour about haunted railroad tracks.
- 0:18
In the recap, children in a school bus were killed by a train on the tracks.
- 0:26
According to the story, a vehicle stopped near the tracks moves by itself and is pushed across.
- 0:44
People are said to put baby powder on the backs of their cars and find children’s handprints in it.
- 0:56
Bell says he has just received a fax and reads it aloud.
- 1:03
The fax says the school-bus, pushing-cars story is true.
- 1:05
The fax says it was on one of the writer’s local news stories that day, with video of the car moving on its own and video of the fingerprints.
- 1:18
The fax is signed [quote withheld] with [quote withheld]
Claims
Each claim is labeled as something the person telling the story noticed (observation), what they made of it (interpretation), something others told them (hearsay), or something the host said (host statement).
- host statement· 0:18
Children in a school bus were killed by a train on the San Antonio tracks.
- host statement· 0:26
A vehicle stopped near the tracks will begin to move by itself and be pushed across.
- host statement· 0:49
People who put baby powder on the backs of their cars later find children’s handprints in it.
- hearsay· 1:03
The fax writer says the school-bus, pushing-cars story is true.
- hearsay· 1:05
The fax writer says a local news broadcast that day showed video of the car moving on its own.
- hearsay· 1:05
The fax writer says the news broadcast had video of the fingerprints.
- host statement· 0:26
The original story was told by a caller the previous hour, who said she had felt the vehicle move herself ([quote withheld]).
Scores, with reasons
An AI model gave these scores using a fixed set of scoring rules (scoring-v1). They are editorial judgments and rough guides, not the odds that anything paranormal happened. How scoring works.
A layered secondhand story. The host recaps another caller’s story and reads a fax that itself cites a television news report. It gives few details that could be checked and has an easy ordinary explanation (a vehicle rolling at a crossing). It is consistent, but the fax simply asserts “it’s true” without hedging.
Evidence for and against
- A concrete mechanism is described: stop near the tracks and the vehicle moves across.
- A named city (San Antonio) and named items of claimed evidence (local news video, fingerprints).
- The fax matches the recap given moments earlier, with no conflicting details.
- No firsthand witness speaks in this part of the broadcast. The story is a recap plus a fax citing a news item.
- No street, date, news station, or named witnesses. The date of the accident is not given.
- The fax writer does not consider ordinary causes for a car moving on a slope or for marks in powder.
- The original caller’s story is outside this part of the broadcast.
- It is unclear whether any video was actually broadcast or what it showed.
- 0:00
- 0:26
- 1:03
What the model inferred (not said on air): Component scoring: source proximity 5 of 25 (hearsay passed through a host and a fax citing the media); specificity 8 of 20; consistency 13 of 20; other explanations 4 of 20 (a slope or tilt at a crossing fits easily and is not addressed); calibration 4 of 15 (a flat claim that it is true). My own judgment, not the transcript, sets the weighting of these components.
Physical evidence is claimed to exist (news video of the car moving and of the fingerprints), but no recording is produced or examined in this part of the broadcast, and no witness speaks on air. This places it in the range for claimed but unexamined physical evidence or claimed witnesses.
Evidence for and against
- The fax says a local news broadcast showed video of the car moving on its own.
- The fax says there was video of the fingerprints.
- Specifics that could in principle be checked: San Antonio, railroad tracks, a same-day local news item.
- Nothing is played or shown on air. The video is only claimed in a fax.
- No second witness speaks. The earlier “lady” is not present in this part of the broadcast.
- No photograph, document, station name, or broadcast date is supplied.
- Whether the claimed video exists, and what it shows, is not stated.
- It is not stated who shot the video or where.
- 1:05
- 1:05
What the model inferred (not said on air): The score rests on treating a claimed television video as [quote withheld] No outside check was done, so the score stays far below the ceiling of 50.
The material includes children killed by a train and children’s handprints on cars, which is eerie. But it comes as a news-confirmation item, not a threatening story: no harm to the living, no intrusion into a safe place, and no lasting dread in this part of the broadcast.
Evidence for and against
- The story’s premise is the deaths of children on the tracks.
- Handprints attributed to children appear in powder on car bodies.
- No threat, attack, or physical contact with any living person is reported.
- The tone and framing here are informational ([quote withheld]).
- The original caller’s fuller description, which might be more vivid, is not in this part of the broadcast.
- 0:18
- 0:49
What the model inferred (not said on air): This judges the effect on the listener only. The deaths and the children’s handprints raise it above a merely neutral report.
The core content, ghost children at a rail crossing pushing stopped cars and leaving handprints in powder, is a widely told legend. Its slightly distinctive feature here is how it is told: a fax arriving moments later and citing same-day television video of both events.
Evidence for and against
- Familiar elements: a fatal school-bus and train accident, cars pushed over tracks, children’s handprints.
- An uncommon framing detail: the confirmation arrives by fax as a news item within the same broadcast.
- No new event or rare combination is described.
- The content is a stock railroad-tracks legend.
- The score compared with the rest of the archive is only provisional for a small pilot archive.
- 1:01
- 1:05
What the model inferred (not said on air): The provisional score reflects stock motifs combined with a mildly unusual way of confirming them. The published score will blend this with how new the story is compared with the rest of the archive.
Scoring confidence 0.55 · analysis confidence 0.60 · checked against outside sources: no · evidence available: Host’s spoken recap of a story told by an earlier caller (not in this part of the broadcast); Fax read on air saying a same-day local television news report included video of a car moving on its own and video of fingerprints; No recording, document, or photograph produced or examined on air. The originality score blends the model's score (35) with how rare the story's details are across the archive. This method is still provisional.
Ordinary explanations to consider
These are possibilities only. None is a finding about what happened.
- A vehicle stopped on or near a rail crossing may roll on its own because of a slope or tilt in the road surface, which would explain the reported self-movement without anything unseen. · The story says the movement happens only when a driver stops the vehicle near the tracks. No other cause is described.
- Living people touching the vehicle could make handprints in powder on a car. Contact, wind, or handling could also disturb the powder. · The reported handprints are read from powder that people visiting the site put on the outside of cars on purpose.
- The [quote withheld] may show faint marks on a vehicle surface that the people filming took for handprints. · The fax only states that video of “the fingerprints” was shown. Nothing in this part of the broadcast describes how the marks were identified or examined.
What stands out
This is not a firsthand story at all. It is a host’s recap followed within moments by a fax claiming that a local television news report, with video of both the moving car and the fingerprints, backs up a story told earlier in the same broadcast. The claim of broadcast video is the distinctive element, not the railroad-tracks motif itself.
Related stories
Matched by shared recurring details, categories, and wording.